Classifications

• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
  • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
  • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
  • H04N19/189—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the adaptation method, adaptation tool or adaptation type used for the adaptive coding
• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
  • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
  • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
  • H04N19/102—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or selection affected or controlled by the adaptive coding
  • H04N19/13—Adaptive entropy coding, e.g. adaptive variable length coding [AVLC] or context adaptive binary arithmetic coding [CABAC]
• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
  • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
  • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
  • H04N19/102—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or selection affected or controlled by the adaptive coding
  • H04N19/119—Adaptive subdivision aspects, e.g. subdivision of a picture into rectangular or non-rectangular coding blocks
• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
  • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
  • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
  • H04N19/134—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or criterion affecting or controlling the adaptive coding
  • H04N19/157—Assigned coding mode, i.e. the coding mode being predefined or preselected to be further used for selection of another element or parameter
• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
  • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
  • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
  • H04N19/134—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or criterion affecting or controlling the adaptive coding
  • H04N19/157—Assigned coding mode, i.e. the coding mode being predefined or preselected to be further used for selection of another element or parameter
  • H04N19/159—Prediction type, e.g. intra-frame, inter-frame or bidirectional frame prediction
• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
  • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
  • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
  • H04N19/169—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding
  • H04N19/17—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object
  • H04N19/174—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object the region being a slice, e.g. a line of blocks or a group of blocks
• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
  • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
  • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
  • H04N19/169—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding
  • H04N19/17—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object
  • H04N19/176—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object the region being a block, e.g. a macroblock
• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
  • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
  • H04N19/46—Embedding additional information in the video signal during the compression process
• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
  • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
  • H04N19/70—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals characterised by syntax aspects related to video coding, e.g. related to compression standards
• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
  • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
  • H04N19/90—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using coding techniques not provided for in groups H04N19/10-H04N19/85, e.g. fractals
  • H04N19/91—Entropy coding, e.g. variable length coding [VLC] or arithmetic coding

Definitions

• Video decoder video encoder, Method for decoding a video content, method for encoding a video content, computer program and video bitstream
• Embodiments according to the invention are related to video decoders, video encoders, methods for decoding a video content, methods for encoding a video content, computer programs and video bitstreams.
• An embodiment according to the invention is related to a probability estimation method for binary arithmetic coding, which can be used, for example, in video encoders, video decoders but also in image encoders, image decoders, audio encoders, audio decoders and the like.
• Arithmetic encoding and decoding has proven to be a valuable tool in the encoding and decoding of audio and video contents. For example, it is possible to exploit known occurrence probabilities of binary values in a binary sequence representing a video or audio content to increase encoding efficiency. In particular arithmetic encoding can deal with varying probabilities of“0”s and“T’s in an efficient manner, and can adapt to changes of the probabilities in a fine-tuned manner.
• tnere is a need tor concepts tor the determination of source statistic values, or range values for the interval sub-division, which provide a good tradeoff between
• An embodiment according to the invention creates a video decoder, wherein the video decoder is configured to decode a plurality of video frames (for example, a sequence of video frames).
• the video decoder is configured to decode a video frame which is subdivided into a set of one or more slices (preferred a plurality of slices), wherein the video decoder is configured to evaluate an slice type information (e.g.“SliceType”) indicating whether a slice is encoded using an independent coding mode (e.g. “Intra”), in which there is no prediction of video content of a current frame on the basis of a video content of a previous frame, or using a single-predictive mode (e.g.“P”) in which there is a prediction of a block of pixels on the basis of one block (e.g. only one block) of pixels of a previous frame (e.g.
• an independent coding mode e.g. “Intra”
• a single-predictive mode e.g.“P”
• the video decoder comprises an arithmetic decoder for providing a decoded binary sequence (for example, describing a transform coefficient of an image content) on the basis of an encoded representation of the binary sequence.
• the arithmetic decoder is configured to determine a first source statistic value (e.g.
• a t (which is, for example, based on a frequency of previously decoded binary values x t-i ,x t-2 ,... ) (and which is, for example, also designated as a“counter variable” or“counter”) using a first window size (e.g. w a ) (which may, for example, be represented by a window size variable), for example, on the basis of a previously decoded binary sequence, and to determine a second source statistic value (e.g. b ( ) (which is based on a frequency of previously decoded binary values x,-i ,x t-2, - - ) using a second window size (e.g.
• the arithmetic decoder is, for example, configured to determine a combined source statistic value (e.g. c t or c' t or p t ) on the basis of the first source statistic value (e.g. a t ) and on the basis of the second source statistic value (e.g. b t ).
• the arithmetic decoder is configured to determine one or more range values for an interval subdivision, which is used for a mapping of the encoded representation of the binary sequence onto the decoded binary sequence, on the basis of the combined source statistic value.
• the arithmetic decoder is configured determine the first window size and the second window size in dependence on the slice type information (for example, to take different values, wherein the first window size and the second window size may, for example lie within a range between 1 and 1 1 , including 1 and 1 1 1
• the video decoder is configured to determine the first window size and the second window size also in dependence on an initialization parameter or flag (for example, a“cabac init flag”) included in the bitstream (for example, one initialization flag per slice), wherein the initialization flag may, optionally also define initialization values for the first source statistic value and/or the second source statistic value.
• an initialization parameter or flag for example, a“cabac init flag” included in the bitstream (for example, one initialization flag per slice)
• the initialization flag may, optionally also define initialization values for the first source statistic value and/or the second source statistic value.
• the video decoder is configured to determine the first window size and the second window size also in dependence on a context model (e.g. in dependence on which type of information is to be decoded, for example whether one or more most significant bits of a transform coefficient or one or more less significant bits of a transform coefficient are to be decoded).
• a context model e.g. in dependence on which type of information is to be decoded, for example whether one or more most significant bits of a transform coefficient or one or more less significant bits of a transform coefficient are to be decoded.
• a pair of window size values, defining the first window size and the second window size may be predefined for each combination of slice type, initialization flag and context model.
• the video decoder is configured to selectively add a predetermined value (e.g.
• the video decoder is configured to selectively add a predetermined value (e.g ((1 « BITS b ) - k b ) » n b or ( ONE b » ?3 ⁇ 4)) to a previously computed instance (e.g. b t ) of the second source statistic value in dependence on a previously decoded binary value (e.g. x ( )) and to subtract a scaled version of the previously computed instance of the second source statistic value (e.g. (b t » ?3 ⁇ 4)), in order to obtain an updated version (e.g. b, +1 ) of the second source statistic value.
• a predetermined value e.g ((1 « BITS b ) - k b ) » n b or ( ONE b » ?3 ⁇ 4)
• a predetermined value e.g ((1 « BITS b ) - k b ) » n b or ( ONE b » ?3
• a is a previously computed version of the first source statistic value, wherein x t is a previously decoded binary value (e.g. binary symbol taking values of 0 or 1 ), wherein n a is a window size information (e.g. having integer values),
• BSTSa (1 « B!TS a ) - k a
• BSTSa is a number of bits used to represent the first source statistic value
• k a is a predetermined number (e.g. an integer number, e.g. 0 or 1 ).
• the video decoder is configured to obtain an updated version h t-M of the second source statistic value according to
• b t +1 b t + x t - ( 0NE b » n b ) - ( b t » n b )
• b t is a previously computed version of the second source statistic value, wherein x t is a previously decoded symbol (e.g. binary symbol taking values of 0 or 1 ), wherein n b is a window size information (e.g. having integer values), wherein
• 0NE b (1 « BITS b ) - k b
• BITS b is a number of bits used to represent the second source statistic value
• k b is a predetermined number (e.g. an integer number, e.g. 0 or 1 ).
• the video decoder is configured to determine the first source statistic value a t+1 according to
• a t +i a t+i ⁇ ( a t+ i ⁇ BITS )
• a L is a previously computed source statistic value
• x t is a previously decoded binary value
• BITS a is a number of bits allocated for the first source statistic value
• b t+ i fy+ i — ( ⁇ t+i » BP S b )
• b is a previously computed source statistic value
• x t is a previously decoded binary value
• BITS b is a number of bits allocated for the first source statistic value
• a t+ 1 a t + x t ( ONE a » n a ) - (a t » n a )
• ONE a represents a probability value equal to 1.
• it may be reasonable to quantize or slightly alter ONE a e.g. by adding or subtracting small quantities). This has virtually no impact on the behavior of the source statistic values. For example, when using
• the video decoder is configured to determine the first source statistic value a t+1 according to
• n is a number of bits allocated for the first source statistic value
• the video decoder is configured to determine the first source statistic value a t+1 according to
• b t is a previously computed source statistic value
• x t is a previously decoded binary value
• BITS b is a number of bits allocated for the first source statistic value
• n b represents the second window size w b
• w b 2 Ub
• “»” is a shift-to-the right operator.
• the video decoder is configured to combine the first source statistic value and the second source statistic value in order to obtain the combined source statistic value.
• the video decoder is configured to combine the first source statistic value and the second source statistic value in order to obtain the combined source statistic value, wherein different weights are associated with the first source statistic value and the second source statistic value (wherein, optionally, the video decoder is configured to change the weights during a decoding process).
• the video decoder is configured to use a different number of bits (e.g. BITS a , BITS b ) for a representation of the first source statistic value (e.g. a t ) and of the second source statistic value (e.g. b,).
• a different number of bits e.g. BITS a , BITS b
• the video decoder is configured to use a different number of bits (e.g. BITS a , BITS b ) for a representation of the first source statistic value (e.g. a t ) and of the second source statistic value (e.g. b,).
• the video decoder is configured to use different window sizes for the first source statistic value (e.g. a t ) and for the second source statistic value (e.g. b t ) and wherein the video decoder is configured to use a comparatively larger number of bits for a representation of the source statistic value having a comparatively larger window size and to use a comparatively smaller number of bits for a representation of the source statistic value having a comparatively smaller window size.
• BITS m m m ⁇ n(B!TS a , BlTS b ) wherein BITS a is a number of bits allocated for a representation of the first source statistic value and wherein BlT8 b is a number of bits allocated for a representation of the second source statistic value.
• the video decoder is configured to determine the combined source statistic value c' t according to
• the video decoder is configured to derive an index value pStateldx designating a table entry of a table comprising pre-calculated range values for the interval-subdivision according to
• BITS tabRes describes a number of entries of the table (or a spacing, in terms of a frequency of the binary value, between entries of the table).
• the table describes, for example, a mapping of probability values within a range from 0 to 0.5 onto pre-calculated range values.
• the video decoder is configured to determine the combined source statistic value c' t according to
• the table describes, for example, a mapping of probability values within a range from 0 to 1 onto pre-calculated range values.
• the video decoder is configured to choose the number of bits (e.g.
• a sum of a total number of bits used for a representation of the first source statistic value (a t ) and of the second source statistic value (b ( ) is constant over different context models (or independent in case of a variation of a number of bits used for the representation of the first source statistic value).
• the video decoder is configured to determine the first window size and the second window size also in dependence on an (for example, dedicated) initialization flag (for example,“wsjfag”) included in the bitstream, and the video decoder is configured to set initialization values for the first source statistic value and/or the second source statistic value in dependence on another initialization flag (for example, a“cabac init flag”) included in the bitstream.
• an initialization flag for example,“wsjfag” included in the bitstream
• a“cabac init flag” included in the bitstream
• the video decoder or the arithmetic decoder is configured to determine the first window size and the second window size also in dependence on a temporal level of a current slice.
• the video decoder or the arithmetic decoder is configured determine the first window size and the second window size also in dependence on a quantization parameter of a current slice.
• the video decoder is configured to adjust the first window size and/or the second window size in dependence on a position (for example, in dependence on how many binary values have already decoded within the current slice, or since a context initialization, using the context model to which the first window size and/or the second window size are associated, or in dependence on a position of a pixel or of a block of pixels within a frame to which a binary value to be decoded is associated).
• the video decoder is configured to set the first window size and/or the second window size to a start value (which is preferably smaller than a norma!
• a binary value associated with a start position e.g. a first row of a block of pixels, or a first column of a block of pixels
• a normal value which is preferably larger than the start value
• the video decoder is configured to change the first window size and/or the second window size within a slice (e.g. when decoding a slice) in response to a signaling flag (e.g a“ctu_ws_flag” flag) signaling that the window size should be changed.
• a signaling flag e.g a“ctu_ws_flag” flag
• the video decoder is configured to evaluate, for a plurality of coding tree units (or even for all coding tree units), a signaling flag indicating whether the window size should be changed, and to increase or decrease at least one of the window sizes in dependence on the signaling flag (for example, to jointly increase or decrease all window sizes, even of different context models, by one in dependence on the signaling flag, keeping in mind not to exceed a maximum range of the window size).
• a signaling flag indicating whether the window size should be changed
• the video decoder is configured to evaluate, for a plurality of coding tree units (or even for all coding tree units), a signaling flag indicating whether the window size should be changed, and to increase or decrease at least one of the window sizes in dependence on the signaling flag (for example, to jointly increase or decrease all window sizes, even of different context models, by one in dependence on the signaling flag, keeping in mind not to exceed a maximum range of the window size).
• the video decoder is configured to derive a state index (e.g. pStateldx) from the combined source statistic value (for example, c t or c' t or Pi), for example according to equation (8) or according to equation (12) or according to equation (14).
• a state index e.g. pStateldx
• the video decoder is configured to derive a value representing a most probable binary value or a least probable binary value (e.g. vaIMPS), for example according to equation (9) or according to equation (13).
• the video decoder is configured to evaluate a table mapping the state index onto the one or more range values for the interval subdivision.
• the video decoder is configured to temporarily (for example, for the decoding of multiple binary values of the binary sequence, not only directly after an initialization) replace the second source statistic value by a constant replacement value, and to combine the first source statistic value (e.g a ( ) with the fixed non zero replacement value in order to obtain the combined source statistic value.
• the video decoder is configured to choose the first window size w a and the second window size w b such that
• Another embodiment according to the invention creates a video decoder, wherein the video decoder is configured to decode a plurality of video frames (for example, a sequence of video frames), and wherein the video decoder is configured to decode a video frame which is subdivided into a set of one or more slices (preferred a plurality of slices).
• a plurality of video frames for example, a sequence of video frames
• the video decoder is configured to decode a video frame which is subdivided into a set of one or more slices (preferred a plurality of slices).
• the video decoder is configured to evaluate a slice type information (e.g.“SliceType”) indicating whether a slice is encoded using an independent coding mode (e.g.“Intra”), in which there is no prediction of video content of a current frame on the basis of a video content of a previous frame, or using a single-predictive mode (e.g.“P”) in which there is a prediction of a block of pixels on the basis of one block of pixels (for example, only one block of pixels) of a previous frame (for example a previously decoded frame), or using a bi-predictive mode (“B”) in which there is a prediction of a block of pixels on the basis of two or more blocks of pixels of one or more previous frames (for example, previously decoded frame), to select a mode of operation for a decoding of a slice.
• a slice type information e.g.“SliceType” indicating whether a slice is encoded using an independent coding mode (e.g.“Intra”)
• the video decoder comprises an arithmetic decoder for providing a decoded binary sequence (for example, describing a transform coefficient of an image content) on the basis of an encoded representation of the binary sequence.
• the arithmetic decoder is configured to determine a first source statistic value (for example, a t ) (which may be based on a frequency of previously decoded binary values c M ,c [-2 ... , and may also be designated as a“counter variable” or“counter”) using a first window size (for example w a), which may, for example, be represented by a window size variable.
• the arithmetic decoder may, for example, determine the first source statistic value on the basis of a previously decoded binary sequence.
• the arithmetic decoder is configured to determine a combined source statistic value (for example, c t or c' t or p t ), wherein the arithmetic decoder is configured to combine the first source statistic value (for example, a t ) with a fixed non-zero value in order to obtain the combined source statistic value, and the arithmetic decoder is configured to determine one or more range values for an interval subdivision, which is used for a mapping of the encoded representation of the binary sequence onto the decoded binary sequence, on the basis of the combined source statistic value.
• a combined source statistic value for example, c t or c' t or p t
• An embodiment according to the invention creates a video encoder, wherein the video encoder is configured to encode a plurality of video frames (for example, a sequence of video frames), wherein the video encoder is configured to encode a video frame which is subdivided into a set of one or more slices (preferably a plurality of slices).
• the video encoder is configured to provide a slice type information (for example, “SliceType”) indicating whether a slice is encoded using an independent coding mode (for example, "Intra”), in which there is no prediction of video content of a current frame on the basis of a video content of a previous frame, or using a single-predictive mode (for example,“P”) in which there is a prediction of a block of pixels on the basis of one block of pixels (for example, only one block of pixels) of a previous frame (for example, a previously encoded frame), or using a bi-predictive mode(for example,“B”) in which there is a prediction of a block of pixels on the basis of two or more blocks of pixels of one or more previous frames (for example, previously encoded frames).
• a slice type information for example, “SliceType” indicating whether a slice is encoded using an independent coding mode (for example, "Intra”), in which there is no prediction of video content of a current frame on the basis of a
• the video encoder comprises an arithmetic encoder for providing an encoded representation of a binary sequence (for example, describing a transform coefficient of an image content) on the basis of the binary sequence (wherein the binary sequence may represent transform coefficients or spectral coefficients, or parameters, or any other information, for example information representing a content of a frame of the video content to be encoded).
• the arithmetic encoder is configured to determine a first source statistic value (a t ), which may, for examplebe based on a frequency of previously encoded binary values x ⁇ x ⁇ and which may, for example, also be designated as a “counter variable” or“counter”, using a first window size (for example, w a ) which may, for example, be represented by a window size variable.
• the video encoder may, for example determine the first source statistic value on the basis of a previously encoded binary sequence.
• the arithmetic encoder or the video encoder may be configured to determine a second source statistic value (for example, b t ), which may, for example, be based on a frequency of previously encoded binary values using a second window size (for example, w b)
• the arithmetic encoder or the video encoder may be configured to determine the second source statistic value, for example, on the basis of the previously encoded binary sequence.
• the arithmetic encoder is configured to determine a combined source statistic value (for example, c t or c' t or p t ) on the basis of the first source statistic value (for example, a t ) and on the basis of the second source statistic value (for example, b ⁇ , and the arithmetic encoder is configured to determine one or more range values for an interval subdivision, which is used for a mapping of the binary sequence onto the encoded representation of the binary sequence, on the basis of the combined source statistic value.
• a combined source statistic value for example, c t or c' t or p t
• the video encoder (or, equivalently, the arithmetic encoder) is configured determine the first window size and the second window size in dependence on the slice type information (for example, two different values, wherein the first window size and the second window size may, for example lie within a range between 1 and 1 1 , including 1 and 1 1 ).
• the video encoder is configured to provide an initialization parameter or flag (for example, a“cabac init flag” flag) and the video encoder is configured to determine the first window size and the second window size also in dependence on the initialization parameter or flag (for example,“cabac init flag”), which is included in a bitstream (for example, a bitstream representing the video frames) by the video encoder.
• the video encoder provides one initialization flag per slice, wherein the initialization flag may, optionally also define initialization values for the first source statistic value and/or the second source statistic value.
• the video encoder is configured to determine the first window size and the second window size also in dependence on a context model (e.g. in dependence on which type of information is to be encoded, for example whether one or more most significant bits of a transform coefficient or one or more less significant bits of a transform coefficient are to be encoded).
• a context model e.g. in dependence on which type of information is to be encoded, for example whether one or more most significant bits of a transform coefficient or one or more less significant bits of a transform coefficient are to be encoded.
• a pair of window size values, defining the first window size and the second window size may be predefined for each combination of slice type, initialization flag and context model.
• the video encoder is configured to selectively add a predetermined value (e.g. ((1 « BITS a ) - k a ) » n a or ONE a » n a ) ) to a previously computed instance (e.g. a t ) of the first source statistic value in dependence on a previously encoded binary value (e.g. x t )) and to subtract a scaled version of the previously computed instance of the first source statistic value (e.g. (a t » n a )), in order to obtain an updated version (a t+1 ) of the first source statistic value.
• a predetermined value e.g. ((1 « BITS a ) - k a ) » n a or ONE a » n a )
• the video encoder is configured to selectively add a predetermined value (e.g. ((1 « BlTS b ) - k b ) » n b or 00NE b » n b )) to a previously computed instance (e.g. b t ) of the second source statistic value in dependence on a previously encoded binary value (e.g. x,)) and to subtract a scaled version of the previously computed instance of the second source statistic value (e.g. (b t » n b )), in order to obtain an updated version (b t+1 ) of the second source statistic value.
• a predetermined value e.g. ((1 « BlTS b ) - k b ) » n b or 00NE b » n b )
• a predetermined value e.g. ((1 « BlTS b ) - k b ) » n b or 00NE b » n
• a t is a previously computed version of the first source statistic value, wherein x t is a previously encoded binary value (e.g. binary symbol taking values of 0 or 1 ), wherein n a is a window size information (e.g. having integer values), wherein
• ONE a (1 « BlTS a ) - k a
• BITSa is a number of bits used to represent the first source statistic value
• k a is a predetermined number (e.g. an integer number, e.g. 0 or 1 ).
• BITS b is a number of bits used to represent the second source statistic value, wherein k b is a predetermined number (e.g. an integer number, e.g. 0 or 1 ).
• b t+1 b [+ i — (b [+1 » B!TS b )
• b t is a previously computed source statistic value
• x t is a previously encoded binary value
• BITS b is a number of bits allocated for the first source statistic value
• b is a previously computed source statistic value
• x t is a previously encoded binary value
• BITS b is a number of bits allocated for the first source statistic value
• the video encoder is configured to determine the first source statistic value a l+1 according to
• a is a previously computed source statistic value
• x is a previously encoded binary value
• BITS a is a number of bits allocated for the first source statistic value
• n a represents the first window size w a
• w a 2 n °
• “»” is a shift-to-the right operator.
• the video encoder may be configured to determine the second source statistic value b t+1 according to
• b! is a previously computed source statistic value
• x t is a previously encoded binary value
• BITS b is a number of bits allocated for the first source statistic value
• n b represents the second window size w b
• w b 2 llb
• “»" is a shift-to-the right operator.
• the video encoder is configured to combine the first source statistic value and the second source statistic value in order to obtain the combined source statistic value, wherein different weights are associated with the first source statistic value and the second source statistic value.
• the video encoder is configured to change the weights during an encoding process.
• the video encoder is configured to use a different number of bits (for example, BITS a , BITS b ) for a representation of the first source statistic value (for example, a t ) and of the second source statistic value (for example, b t ).
• the video encoder is configured to use different window sizes for the first source statistic value (for example, a,) and for the second source statistic value (for example, b t ), and the video encoder is configured to use a comparatively larger number of bits for a representation of the source statistic value having a comparatively larger window size and to use a comparatively smaller number of bits for a representation of the source statistic value having a comparatively smaller window size.
• B!TS i m min (B lTS , B lTS ), and wherein BITSa is a number of bits allocated for a representation of the first source statistic value and wherein BITSb is a number of bits allocated for a representation of the second source statistic value.
• the video encoder is configured to determine the combined source statistic value c' t according to
• the video encoder is configured to derive an index value pStateldx designating a table entry of a table comprising pre-calculated range values for the interval-subdivision according to
• the table describes, for example, a mapping of probability values within a range from 0 to 0.5 onto pre-calculated range values.
• an index value pStateldx designating a table entry of a table comprising pre-calculated range values for the interval-subdivision according to pStateldx— c’ t » ( BITS, nin — BlTS [abRes ), wherein til I 3 ⁇ 4 ai ,F!P, describes a number ot entries ot the table (or a spacing, in terms of a frequency of the binary value, between entries of the table)
• the table describes for example, a mappinrg of probability values within a range from 0 to 1 onto pre-calculated range values.
• a sum of a total number of bits used for a representation of the first source statistic value (for example, a t ) and of the second source statistic value (for example, b t ) is constant over different context models (or independent in case of a variation of a number of bits used for the representation of the first source statistic value).
• the video encoder is configured to include an initialization flag (for example, a“ws_flag” or a window size flag) into the bitstream which determines (possibly in addition with other configuration information, like slice type) the first window size and the second window size, and the video encoder is configured to include another initialization parameter or flag (for example, a“cabac init flag”) into the bitstream which determines initialization values for the first source statistic value and/or the second source statistic value.
• an initialization flag for example, a“ws_flag” or a window size flag
• another initialization parameter or flag for example, a“cabac init flag
• the video encoder is configured determine the first window size and the second window size also in dependence on a temporal level of a current slice. Alternatively or in addition, the video encoder is configured determine the first window size and the second window size also in dependence on a quantization parameter of a current slice.
• the video encoder is configured to adjust the first window size and/or the second window size in dependence on a position (for example, in dependence on how many binary values have already encoded within the current slice, or since a context initialization, using the context model to which the first window size and/or the second window size are associated, or in dependence on a position of a pixel or of a block of pixels within a frame to which a binary value to be encoded is associated).
• the video encoder is configured to set the first window size and/or the second window size to a start value (which is preferably smaller than a normal value) when encoding a binary value associated with a start position (e.g. a first row of a Block of pixels, or a first column of a block of pixels), and to set the first window size and/or the second window size to a normal value (which is preferably larger than the start value) when encoding a binary value which is associated with a position which is distanced from the start position by at least a predetermined position.
• a start value which is preferably smaller than a normal value
• a normal value which is preferably larger than the start value
• the video encoder is configured to include signaling flag (for example a “ctu_ws_flag” or a window size change signaling flag) into the bitstream signaling that the window size should be changed within a slice.
• signaling flag for example a “ctu_ws_flag” or a window size change signaling flag
• the video encoder is configured to include into the bitstream the signaling flag indicating whether the window size should be changed, for a plurality of coding tree units (or even for all coding tree units), wherein the signaling flag indicates whether to increase or decrease at least one of the window sizes (for example, to jointly increase or decrease all window sizes, even of different context models, by one in dependence on the signaling flag, keeping in mind not to exceed a maximum range of the window size).
• the video encoder may only increase or decrease window sizes of some of the context models.
• the video encoder is configured to derive a state index (for example, pStateldx) from the combined source statistic value (for example, c t or c' t or p t ), for example, according to equation (8) or according to equation (12) or according to equation (14)].
• the video encoder may be configured to derive a value representing a most probable binary value or a least probable binary value (e.g. vaIMPS), for example according to equation (9) or according to equation (13).
• the video encoder may also be- configured to evaluate a table mappinci the stale index onto the one or more range values for the in!emai subdivision
• the video encoder may be configured to temporarily (for example, for the encoding of multiple binary values of the binary sequence, not only directly after an initialization) replace the second source statistic value by a constant replacement value, and to combine the first source statistic value (e.g. a t ) with the fixed non zero replacement value in order to obtain the combined source statistic value.
• the video encoder is configured to choose the first window size w a and the second window size w b such that
• An embodiment according to the invention creates a video encoder, wherein the video encoder is configured to encode a plurality of video frames (for example, a sequence of video frames),
• the video encoder is configured to encode a video frame which is subdivided into a set of one or more slices (preferred a plurality of slices).
• the video encoder is configured to provide an slice type information (for example, “SliceType”) indicating whether a slice is encoded using an independent coding mode (for example, “Intra”), in which there is no prediction of video content of a current frame on the basis of a video content of a previous frame, or using a single-predictive mode (for example,“P”) in which there is a prediction of a block of pixels on the basis of one block of pixels (for example, only one block of pixels) of a previous frame (for example, a previously encoded frame]), or using a bi-predictive mode (for examp!e,“B”) in which there is a prediction of a block of pixels on the basis of two or more blocks of pixels of one or more previous frames (for example, previously encoded frames).
• an independent coding mode for example, “Intra”
• the video encoder comprises an arithmetic encoder for providing an encoded representation of a binary sequence (for example, describing a transform coefficient of an image content) on the basis of the binary sequence (wherein the binary sequence may represent transform coefficients or spectral coefficients, parameters or any other information).
• the arithmetic encoder is configured to determine a first source statistic value (for example a ( ), which is, for example, based on a frequency of previously encoded binary values x t _i,x t-2 ,...and which may also be designated as a "counter variable” or as a counter, using a first window size (for example, w a ) which may, for example, be represented by a window size variable.
• the arithmetic encoder may be configured to determine a first source statistic value, for example, on the basis of a previously encoded binary sequence.
• the arithmetic encoder is configured to determine a combined source statistic value (for example, c t or c' t or p t ), wherein the arithmetic encoder is configured to combine the first source statistic value (for example, at) with a fixed non-zero value in order to obtain the combined source statistic value, and the arithmetic encoder is configured to determine one or more range values for an interval subdivision, which is used for a mapping of the binary sequence onto the encoded representation of the binary sequence, on the basis of the combined source statistic value.
• the embodiment can optionally be supplemented by any of the features, functionalities and details of the embodiments described before, both individually and taken in combination.
• An embodiment according to the invention creates a method for decoding a video content, wherein the method comprises decoding a plurality of video frames (for example, a sequence of video frames), and wherein the method comprises decoding a video frame which is subdivided into a set of one or more slices (preferred a plurality of slices).
• the method comprises evaluating a slice type information (for example, "SliceType”) indicating whether a slice is encoded using an independent coding mode (for example,“Intra”), in which there is no prediction of video content of a current frame on the basis of a video content of a previous frame, or using a single-predictive mode (for example,“P”) in which there is a prediction of a block of pixels on the basis of one block of pixels (or only one block of pixels) of a previous frame (for example, previously decoded frame), or using a bi-predictive mode (for example, “B”) in which there is a prediction of a block of pixels on the basis of two blocks of pixels of one or more previous frames (for example, previously decoded frames), to select a mode of operation for a decoding of a slice.
• an independent coding mode for example,“Intra”
• P single-predictive mode
• B bi-predictive mode
• the method comprises providing a decoded binary sequence [for example, describing a transform coefficient of an image content] on the basis of an encoded representation of the binary sequence, and the method comprises determining a first source statistic value (for example, a t ), which may, for example, be based on a frequency of previously decoded binary values x M , x l-2 , ... , and which may, for example, also be designated as a “counter variable” or“counter”, using a first window size (for example, w a ) which may, for example, be represented by a window size variable.
• the first source statistic value may, for example, be determined, on the basis of a previously decoded binary sequence.
• the method also comprises determining a second source statistic value (for example, b t ), which may, for example, be based on a frequency of previously decoded binary values c M ,c (- 2, . . . , using a second window size (for example, w b ).
• the second source statistic value may, for example, be determined on the basis of the previously decoded binary sequence.
• the method also comprises determining a combined source statistic value (for example c t or c' t or p t ) on the basis of the first source statistic value (for example, a t ) and on the basis of the second source statistic value (for example, b,).
• the method also comprises determining one or more range values for an interval subdivision, which is used for a mapping of the encoded representation of the binary sequence onto the decoded binary sequence, on the basis of the combined source statistic value.
• An embodiment according to the invention creates a method for decoding a video content, wherein the method comprises decoding a plurality of video frames (for example a sequence of video frames), and wherein the method comprises decoding a video frame which is subdivided into a set of one or more slices (preferred a plurality of slices).
• the method comprises evaluating a slice type information (for example,“SliceType”) indicating whether a slice is encoded using an independent coding mode (for example,“Intra”), in which there is no prediction of video content of a current frame on the basis of a video content of a previous frame, or using a single-predictive mode (for example,“P”) in which there is a prediction of a block of pixels on the basis of one block of pixels (for example, only one block of pixels) of a previous frame (for example, previously decoded frame), or using a bi-predictive mode (for example,“B”) in which there is a prediction of a block of pixels on the basis of two or more blocks of pixels of one or more previous frames (for example, previously decoded frames), to select a mode of operation for a decoding of a slice.
• an independent coding mode for example,“Intra”
• P single-predictive mode
• B bi-predictive mode
• the method comprises providing a decoded binary sequence (for example, describing a transform coefficient of an image content) on the basis of an encoded representation of the binary sequence.
• the method comprises determining a first source statistic value (for example, a t ), which may, for example, be based on a frequency of previously decoded binary values c M ,c , 2> ... and which may also be designated as a“counter variable”,“counter” using a first window size (for example, w a ) which may, for example, be represented by a window size variable.
• the first source statistic value may be determined on the basis of a previously decoded binary sequence.
• the method comprises determining a combined source statistic value (for example, c t or c' t or p t ), wherein the first source statistic value (for example a ( ) is combined with a fixed non-zero value in order to obtain the combined source statistic value.
• the method further comprises determining one or more range values for an interval subdivision, which is used for a mapping of the encoded representation of the binary sequence onto the decoded binary sequence, on the basis of the combined source statistic value.
• An embodiment creates a method for encoding a video content, wherein the method comprises encoding a plurality of video frames (for example, a sequence of video frames), and wherein the method comprises encoding a video frame which is subdivided into a set of one or more slices (preferred a plurality of slices).
• the method comprises providing slice type information (for example, “SliceType”) indicating whether a slice is encoded using an independent coding mode (for example, “Intra”), in which there is no prediction of video content of a current frame on the basis of a video content of a previous frame, or using a single-predictive mode (for example,“P”) in which there is a prediction of a block of pixels on the basis of one block of pixels (for example, only one block of pixels) of a previous frame (for example, a previously encoded frame), or using a bi-predictive mode (for example,“B”) in which there is a prediction of a block of pixels on the basis of two blocks of pixels of one or more previous frames (for example, previously encoded frames).
• the method comprises providing an encoded representation of a binary sequence (for example, describing a transform coefficient of an image content) on the basis of the binary sequence (wherein the binary sequence may represent transform coefficients or spectral coefficients, parameters or any other information).
• the method comprises determining a first source statistic value (for example, a t ), which may, for example, be based on a frequency of previously encoded binary values c ,c,_ 2, ... and which may, for example, be designated as a“counter variable” or“counter”, using a first window size (for example, w a ) which may, for example, be represented by a window size variable.
• the first source statistic value may, for example, be determined on the basis of a previously encoded binary sequence.
• the method also comprises determining a second source statistic value (for example, b t ), which may, for example, be based on a frequency of previously encoded binary values C M , CI-2, ⁇ ⁇ ⁇ , using a second window size (for example, w b ).
• the second source statistic value may, for example, be determined on the basis of the previously encoded binary sequence.
• the method also comprises determining a combined source statistic value (for example, c t or c' r or p ( ) on the basis of the first source statistic value (for example, a ( ) and on the basis of the second source statistic value (for example, b t ).
• the method comprises determining one or more range values tor an interval subdivision, which is used for a mapping of the binary sequence onto the encoded representation of the binary sequence, on the basis of the combined source statistic value.
• An embodiment creates a method for encoding a video content, wherein the method comprises encoding a plurality of video frames (for example, a sequence of video frames), wherein the method comprises encoding a video frame which is subdivided into a set of one or more slices (preferred a plurality of slices).
• the method comprises providing a slice type information (fo example, “SliceType”) indicating whether a slice is encoded using an independent coding mode (for example, “Intra”), in which there is no prediction of video content of a current frame on the basis of a video content of a previous frame, or using a single-predictive mode (for example,“P”) in which there is a prediction of a block of pixels on the basis of one block of pixels (for example, only one block of pixels) of a previous frame (for example, a previously encoded frame), or using a bi-predictive mode (for example, "B”) in which there is a prediction of a block of pixels on the basis of two or more blocks of pixels of one or more previous frames (for example, previously encoded frames).
• an independent coding mode for example, “Intra”
• P single-predictive mode
• B bi-predictive mode
• the method comprises providing an encoded representation of a binary sequence (for example, describing a transform coefficient of an image content) on the basis of the binary sequence (wherein the binary sequence may represent transform coefficients, or spectral coefficients, parameters or any other information).
• the method comprises determining a first source statistic value (for example, a , which may, for example, be based on a frequency of previously encoded binary values x M ,Xi_ 2 , . . . and which may, for example, be designated as a “counter variable” or“counter”, using a first window size (for example, w a ) which may be represented by a window size variable.
• the first source statistic value may, for example be determined on the basis of a previously encoded binary sequence.
• the method comprises determining a combined source statistic value (for example, c t or c' t or p t ), wherein the first source statistic value (for example, a ( ) is combined with a fixed non-zero value in order to obtain the combined source statistic value.
• the method comprises determining one or more range values for an interval subdivision, which is used for a mapping of the binary sequence onto the encoded representation of the binary sequence, on the basis of the combined source statistic value.
• Another embodiment according to the invention creates a computer program for performing at least one of the methods described herein when the computer program runs on a computer.
• An embodiment according to the present invention creates a video bitstream comprising an encoded representation of the binary sequence, which represents the video content, and a configuration information describing a configuration of a video decoder to be used to reconstruct the video content on the basis of the encoded representation of the binary sequence.
• the configuration information comprises a window size information (for example, “ws_flag”) describing a window size to be used by a video decoder to determine a first source statistic value (for example, a t ) which is based on a frequency of previously decoded binary values and which is used to determine one or more range values for an interval subdivision, wherein the one or more range values are used for a mapping of the encoded representation of the binary sequence onto the decoded binary sequence.
• ws_flag describing a window size to be used by a video decoder to determine a first source statistic value (for example, a t ) which is based on a frequency of previously decoded binary values and which is used to determine one or more range
• the configuration information comprises a window size information (for example, “ws_flag”) describing a window size to be used by a video decoder to determine a second source statistic value (for example, b t ) which is based on a frequency of previously decoded binary values and which is used to determine one or more range values for an interval subdivision.
• a window size information for example, “ws_flag”
• b t second source statistic value
• the video bitstream also comprises an initialization value information describing an initialization value of the first source statistic value and/or of the second source statistic value.
• the video bitstream further comprises a window size change flag (for example, ctu_ws_flag) which indicates that the window size should be increased or decreased (for example, within a slice).
• a window size change flag for example, ctu_ws_flag
• the video stream can be supplemented by any of the features and details described herein both individually and in combination.
• window size an estimation parameter may optionally be used in any of the embodiments.
• first estimation parameter can optionally be used in any of the embodiments.
• second estimation parameter can optionally be used in any of the embodiments.
• the first estimation parameter may be a first window size
• the second estimation parameter may be a second window size
• the first estimation parameter may, alternatively, be a first inertia parameter which determines a speed at which the first source statistic value follows a change of a frequency of decoded binary symbols in the decoded binary sequence (or of previously encoded binary values, in the case of an encoder), and the second estimation parameter may alternatively be a second inertia parameter which determines a speed at which the second source statistic value follows a change of a frequency of decoded binary symbols in the decoded binary sequence (or of previously encoded binary values, in the case of an encoder).
• the video decoder may be configured to determine a first source statistic value using a table lookup, wherein a table entry is selected in dependence on a previously determined source statistic value, a previously decoded binary value and the first estimation parameter.
• the video decoder is configured to determine a second source statistic value using a table lookup, wherein a table entry is selected in dependence on a previously determined source statistic value, a previously decoded binary value and the second estimation parameter.
• a similar concept can optionally be used in a video encoder, wherein a previously encoded binary value may take the role of the previously decoded binary value.
• Fla. f shows a block schematic diagram of a video decoder, according to an embodiment of the present invention.
• Fig. 2 shows a block schematic diagram of a video decoder, according to another embodiment of the present invention.
• Fig. 3 shows a block schematic diagram of a video encoder, according to an embodiment of the invention
• Table 1 shows a representation of a bitmask of an 8-bit variable ws pair ;
• Table 2 shows a graphic representation of an assignment of ws pair values
• Table 3 shows a graphic representation of an assignment of ws pair values dependent on ws_flag
• Fig. 4 shows a flowchart of a method for video decoding, according to an embodiment of the present invention
• Fig. 5 shows a flowchart of a method video encoding, according to an embodiment of the present invention.
• Fig. 6 shows a graphic representation of a bit stream, according to an embodiment of the invention.
• Fig 1 shows a block schematic diagram of a video decoder 100, according to an embodiment of the invention.
• the video decoder 100 is configured to receive an encoded video information and to provide, on the basis thereof, a decoded video information (or decoded video content).
• the encoded video information 1 10 (which may be considered as a video bitstream) may, for example, comprise a slice-type information and may also comprise an encoded representation of a binary sequence.
• the encoded video information 1 10 may comprise additional information which, however, is not essential for the present invention.
• the video decoder is configured to decode a plurality of video frames (for example, a sequence of video frames), and the video decoder is, in particular, configured to decode a video frame which is subdivided into a set of one or more slices (preferably, into a plurality of slices).
• the video decoder is also configured to evaluate the slice type information, which is included in the encoded video information 1 10, and which indicates whether a slice is encoded using an independent coding mode, in which there is no prediction of video content of a current frame on the basis of a video content of a previous frame, or using a single-predictive mode, in which there is a prediction of a block of pixels on the basis of one block of pixels of a previous frame, or using a bi-predictive mode in which there is a prediction of a block of pixels on the basis of two or more blocks of pixels of one or more previous frames, to select a mode of operation for a decoding of a slice (which may, for example, be performed by a "video reconstruction” block 180).
• an independent coding mode in which there is no prediction of video content of a current frame on the basis of a video content of a previous frame
• a single-predictive mode in which there is a prediction of a block of pixels on the basis of one block of pixels of a previous frame
• the video decoder 100 comprises an arithmetic decoder 120 which is, for example, configured to provide a decoded binary sequence 122 (for use by the“video reconstruction” block) on the basis of the encoded representation of the binary sequence, which is included in the encoded video information 1 10.
• the arithmetic decoder preferably comprises a first source static value determination 130 and a second source statistic value determination 140. Accordingly, the arithmetic decoder 120 is configured to determine a first source statistic value 132 using a first window size and to determine a second source statistic value 142 using a second window size.
• the arithmetic decoder also preferably comprises a combiner 150.
• the arithmetic decoder is configured to determine a combined sourced statistic value 152 on the basis of the first source static value and on the basis of the second source statistic value.
• the arithmetic decoder 120 preferably comprises a range value determination 160.
• the arithmetic decoder may be configured to determine one or more range values for an interval subdivision, which is used for a mapping of the encoded representation of the binary sequence (which is included in the encoded video information 1 10) onto the decoded binary sequence 122 (which is used by the video reconstruction block 180), on the basis of the combined source statistic value 152.
• the arithmetic decoder 120 also comprises an arithmetic decoding core 170 (which may, for example, be a block or a unit), which receives the one or more range values 162 from the range value determination 160, and which uses the range values to derive the decoded binary sequence 122 from the encoded binary sequence included in the encoded video information 1 10.
• an arithmetic decoding core 170 (which may, for example, be a block or a unit), which receives the one or more range values 162 from the range value determination 160, and which uses the range values to derive the decoded binary sequence 122 from the encoded binary sequence included in the encoded video information 1 10.
• the video decoder may, for example, comprise a video reconstruction block (or unit) 180, which receives the decoded binary sequence 122 and which provides the decoded video content 1 12 on the basis of the decoded binary sequence 122 (possibly considering additional control information, like the slice-type information).
• a video reconstruction block (or unit) 180 which receives the decoded binary sequence 122 and which provides the decoded video content 1 12 on the basis of the decoded binary sequence 122 (possibly considering additional control information, like the slice-type information).
• the arithmetic decoder 100 receives the encoded video information 110 and performs an arithmetic decoding of an encoded representation of a binary sequence, to derive the decoded binary sequence 122.
• the arithmetic decoding exploits knowledge about probabilities of binary values in the decoded binary sequence 122. This knowledge about the probability (or estimated probability) of binary values within the decoded binary sequence 122 is considered by the arithmetic decoding core 170 by relying on range values 162 which define an interval subdivision. Briefly speaking, the arithmetic decoding core can use the range values 162 to define different intervals (for example, between zero and one, or over a range of integer values).
• the arithmetic decoding core can, for example, interpret the encoded representation of the binary sequence as a representation of a number, which lies in one of the intervals defined using the range values. By recognizing in which of the intervals the number represented by the encoded representation of the binary sequence lies, the arithmetic decoding core 170 can conclude which bit or which bit sequence has been encoded using I be encoded representation of the binary sequence.
• arithmetic decoding core 170 should only be considered a very brief and generic explanation. Details regarding the arithmetic decoding core can, for example, be seen in the standards H.264 and H.265. However, different concepts (for the operation of an arithmetic decoding core) can also be seen in the literature, and the details of the arithmetic decoding core are not of particular relevance for the present invention.
• the arithmetic decoder 120 determines two source statistic values 132, 142 using different window size (wherein the“window size" defines a degree of smoothing over a plurality of decoded binary values of the decoded binary sequence 122). Moreover, to increase the reliability of the range values provided to the arithmetic decoding core 170, the first source statistic value 132 and the second source statistic value 142 are combined into the combined source statistic value 152.
• the video decoder 100 provides a high efficiency, because the range values used by the arithmetic decoding core 170 are well-adapted to the actual probability of bit values (for example, of bit values“0” and“1” within the decoded binary sequence 122).
• the video decoder 100 can also be modified.
• the second source statistic value determination 140 can be replaced by the provision of a fixed value (which may be independent from the decoded binary sequence, but which may depend on one or more parameters).
• the arithmetic decoder is configured to combine the first source statistic value 132 with the fixed non-zero value in order to obtain the combined source statistic value 152. It has been found that such a simplification brings good results in some cases and may, for example, avoid inappropriately strong variations of the combined source statistic value. In other words, by introducing a fixed contribution into the determination of the combined source statistic value, it can be achieved that the combined source statistic value can no longer deviate too strongly from this fixed value. Accordingly, some“hindsight” into the statistics of the decoded binary sequence can be used lo avoid a strong degradation of the coding efficiency ,f, by chance, there is a longer sequence of identical bit values within the decoded binary sequence 122.
• the functionality of the arithmetic decoder (and of the individual blocks of the arithmetic decoder) can, generally, also be considered as a functionality of the video decoder in its entirety.
• the functionalities described herein as functionalities of the arithmetic decoder can also be performed by other blocks of the video decoder.
• Fig. 1 the video decoder 100 according to Fig. 1 can be supplemented by any of the features, functionalities and details described herein, both individually and taken in combination.
• Fig. 2 shows a block schematic diagram of a video decoder 200, according to an embodiment of the invention.
• the video decoder 200 is configured to receive an encoded video information 210 (for example, a video bitstream) and to provide, on the basis thereof, a decoded video content 212 (for example, a sequence of video frames).
• the encoded video information 210 may, for example, comprise a slice type information, as described herein.
• the encoded video information 210 may further comprise configuration information, which can also be considered as control information.
• the encoded video information 210 may comprise an encoded representation of a binary sequence.
• Fig. 2 two main blocks of the video decoder 200 are shown, namely an arithmetic decoder 220 and video reconstruction 280.
• the arithmetic decoder 220 receives the encoded representation 21 1 of the binary sequence.
• the arithmetic decoder (or a control block which may be outside of the arithmetic decoder) also receives the slice type information and the configuration information (or control information).
• the arithmetic decoder 220 provides a decoded binary sequence 222 to the video reconstruction 280 on the basis of the encoded representation 21 1 of the binary sequence and taking into consideration some or all of the slice type information and the configuration information or control information.
• the arithmetic decoded comprises an arithmetic decoding core 270, which receives the encoded representation 211 of the binary sequence and which provides the decoded binary sequence 222.
• the arithmetic decoding core determines which bit values of the decoded binary sequence 222 are represented by the encoded representation 21 1 of the binary sequence.
• the arithmetic decoding core 270 checks in which interval of a range of numbers a number represented by the encoded representation 211 of the binary sequence lies.
• interval out of a plurality of (at least two) intervals, a number represented by the encoded representation 211 of the binary sequence lies, a certain bit value, or a group of bit values, of the decoded binary sequence 222 is recognized.
• the arithmetic decoding core receives an information about the intervals, which typically corresponds to some degree with probabilities of bit values.
• the arithmetic decoding core 270 receives “range values” 262 which are used for the interval subdivision (i.e., range values 162 which serve to define intervals of the number range to be used by the arithmetic decoding core 270).
• range values 162 which serve to define intervals of the number range to be used by the arithmetic decoding core 270.
• the arithmetic decoding core 270 may, for example, be similar or identical to an arithmetic decoding core used in a video encoder/decoder according to standard H.264 or in a video encoder/decoder according to standard H.265.
• different approaches to realize the arithmetic decoding core 270 could also be used.
• the arithmetic decoder 220 derives these range values 262 from previously decoded binary values of the decoded binary sequence 222, taking into consideration some control-information which defines parameters like, for example, initialization values, "window sizes”, “window size adaptation ' ', and the like.
• the first source statistic value determination block 230 receives one or more previously decoded binary values of the decoded binary sequence 222 (also designated with x t ), and provides, on the basis thereof, a first sources statistic value 232.
• the first source statistic value determination block may, for example, receive some configuration information, like a constant or variable BITS a , defining a number of bits used to represent the source statistic value 232, and a constant or variable n a which defines a“window size” to be used by the source statistic value determination block 230.
• the first source statistic value determination block 230 may recursively determine the first source statistic value 232, wherein the window size n a determines a weighting of a most recently decoded binary value of the decoded binary sequence 222 in the determination of the first source statistic value 232.
• the functionality of the first source statistic value determination block 230 is, for example, similar to the formation of a sliding average having a certain window size, except for the fact that a recursive algorithm is used which introduces some “infinite impulse response” characteristic.
• the first source statistic value 132 does not exactly represent a result of a sliding window summation operation or of a sliding window averaging operation, but rather should be considered as a“virtual sliding window” operation, since the result is very similar.
• the second source static value determination block 240 performs a similar operation when compared to the first source statistic value determination block 230.
• the second source statistic value determination block 240 typically uses different parameters (for example, different window length n b and/or a different bit number parameter BITS b ), and consequently provides a second source statistic value 242, which is typically different from the first source statistic value 232.
• one of the source statistic values 232, 242 may be a short term (or shorter term) average source statistic value and one of the source statistic values 232 242 may be a lono term (or lonoGr-torm ) avcracic source statistic value
• the source statistic value determination blocks 230, 240 may, for example, perform a functionality as defined by equations (3) and (4) discussed in detail below.
• the source statistic value determination blocks 230, 240 may also perform the functionality as defined by equation (5) discussed below.
• different computation rules could also be used in the source statistic value determination blocks 230, 240 in some embodiments.
• the arithmetic decoder 220 further comprises a combined source statistic value determination block (or unit) 250, which is configured to receive the first source statistic value 232 and the second source statistic value 242.
• the source statistic value combination block 250 provides a combined source statistic value 252 on the basis thereof.
• the source statistic value combination block 250 may form a sum or an average of the first source statistic value 232 and of the second source statistic value 242, to thereby obtain the combined source statistic value 252.
• the source statistic value combination block 250 could also apply different weightings to the first source statistic value 232 and to the second source statistic value 242 when deriving the combined source statistic value 252, wherein the different weightings may even vary within a slice or between different slices.
• the source statistic value combination block 250 may perform the functionality as defined by equation (6) mentioned below or as defined by equations (10) and (1 1 ) mentioned below or as defined by equation (16) mentioned below. However, a variation of this functionality is also possible.
• the source statistic value combination block 250 combines only one of the first statistic values with a fixed value, to thereby obtain the combined source statistic value 252.
• Such a concept may be advantageous to avoid that the combined source statistic value 252 deviates by too much from an expected probability of binary values within the decoded binary sequence 222.
• the arithmetic decoder 220 is configured to derive the range values 262 for the interval combined source statistic value 252.
• This processing step may, for example, be considered as“range value determination".
• the range value determination may comprise an optional value processing 286, which receives the combined source statistic value 252 and provides, on the basis thereof, a probability value or a state index value.
• the value processing 288 may, for example, map the range of values of the combined source statistic value 252 onto a range between 0 and 1 , or onto a range between 0 and 0.5, or onto an integer index value.
• the value processing 268 may be performed according to equation (7) described below, or according to equations (8), (9) described below, or according to equations (12), (13) described below, or according to equation (14) described below.
• the value processing 266 may provide information 267, which may be a binary information that indicates whether it is more likely that the next decoded value (for example, of the decoded binary sequence 222) takes a value of“1” or takes a value“0”.
• the arithmetic decoder (or the range value determination) may comprise a mapping table 269.
• the mapping table 269 may, for example, receive an index value (for example, pStateldx) which designates a table entry. Accordingly, the mapping table 269 may provide one or more range values 262 corresponding to said table entry designated by the index value (for example pStateldx).
• one or more range values for the interval subdivision may be provided on the basis of the combined source static value 252.
• the mapping table 269 may, for example, have the same structure like a mapping table described in the standard H.264 or in the standard H.265. However, optionally, the contents of the mapping table may be adapted to the specific details of the video decoder. In particular, entries of the mapping table can be adapted to the statistic properties expected in the specific video decoder.
• the arithmetic decoder also comprises a control block 290 which may receive a control information or configuration information and which may adjust the parameters used for the provision of the range values (and possibly also other parameters, for example additional parameters used by the arithmetic decoding core 270) on the basis thereof.
• the control block 290 may receive one or more of the slice type information, a "cabac mil flag " , a "wsjlacf and a " ctu_ws_ flag", which may be included in the encoded video information 210.
• control 290 may, for example, adjust the window size parameters n a , n b and the bit size parameters BITSa, BITSb in dependence on the control information.
• control block 290 may also a consider a current context model.
• this issue it should be noted that, for each bit (or group of bits) of the decoded binary sequence 222 to be decoded, it may be decided which context model is to be used. For example, the decision which context model is to be used may be based on the fact which type of information (decoding parameter, transform coefficient, etc.) is represented by the respective bit (or group of bits).
• control block 290 may be configured to recognize the syntax of the decoded binary sequence 222, to thereby recognize which syntax element (or which part of a syntax element, for example, a most significant bit or a least significant bit, or the like) is to be decoded next Accordingly, a selection between different context models may be made. Also, it should be noted that the window size parameters and/or the bit size parameters and/or any other parameters may be chosen dependent on the context model. Furthermore, it should also be noted that the source statistic values 232,242 or the combined source statistic value 252 may be associated with a specific context model, such that different source statistic values or combined source statistic values may be available for different context models.
• a source statistic value associated with a certain context model may be selectively provided on the basis of decoded binary values of the decoded binary sequence 222 which have been decoded using the respective context model.
• a separate, independent processing, and a separate (possibly independent) decision regarding the parameters n a , n b , BITS a , BITS b, etc. may be made for different context models.
• control unit 290 may determine the parameters n a , n b, BITS a , BITS b, for example, in accordance with the mechanisms described below.
• the window size parameters n a, n b may be chosen in dependence on the slice type information, and/or in dependence on the cabac init flag, and/or in dependence on the ws__flag and/or in dependence on the ctu_ws_flag.
• bit size parameters BITSa, BITSb may be chosen in dependence on some configuration information in some embodiments. However, in other embodiments, the bit size parameters may be fixed.
• the video reconstruction block 280 typically receives the decoded binary sequence 222 and also at least some elements of the configuration information.
• the video reconstruction 280 may reconstruct integer parameters and/or floating point parameters and/or image data on the basis of the decoded binary sequence 222.
• the video reconstruction block 280 reconstructs information used for the reconstruction of a video frame from the decoded binary sequence 222. Then, the video reconstruction block may generate image information on the basis of the reconstructed information (derived from the decoded binary sequence 222).
• the video reconstruction 280 may comprise the same functionality as describe in the standard H.264 or in the standard H.265. However, other approaches may also be used for the video reconstruction which are adapted to provide a decoded video content on the basis of a decoded binary sequence (and, possibly, additional configuration information or control information). Accordingly, the video reconstruction 280 provides a decoded video content 212, which may take the form of as sequence of video frames.
• Fig. 3 shows a block schematic diagram of a video encoder 300, according to an embodiment of the present invention.
• the video encoder 300 is configured to receive a video content 310 and to provide, on the basis thereof, an encoded video information (for example, a video bitstream) 312.
• an encoded video information for example, a video bitstream
• the video encoder 300 comprises a video binary sequence provision block 380, which is configured to receive the video content 310 to provide, on the basis thereof, a binary sequence 322 which represents the video content 310.
• the video binary sequence provision 380 may be performed like in a video encoder according to the standard H.264 or the like in a video encoder according to the standard H.265.
• different approaches for the provision of a video binary sequence may also be used.
• the video binary sequence provision 380 may, for example, comprise a mapping of integer-valued or floating-point parameters or image data (like, for example, transform coefficients) onto a sequence of binary values.
• an arithmetic encoder 320 may receive the binary sequence 322 and provide, on the basis thereof, an encoded representation 311 of the binary sequence 322.
• the arithmetic encoder 320 is configured to exploit knowledge (or information) about probabilities of binary values within the binary sequence 322 to provide an efficiently compressed representation of the binary sequence 322 (namely the encoded representation 31 1 of the binary sequence).
• the arithmetic encoder 320 may, for example, comprise an arithmetic encoding core 370, which receives the binary sequence 322 and which provides, on the basis thereof, the encoded representation 31 1 of the binary sequence.
• the arithmetic encoding core 370 may, for example, require some knowledge about probabilities of bits (or groups of bits) in the binary sequence 322, in order to be able to find a proper codeword (which is included into the encoded representation 31 1 ) representing the bit or group of bits.
• the arithmetic encoding core 370 receives range values 362 which describe a subdivision of intervals.
• range values 362 may describe a subdivision of intervals which is to be used by a video decoder 200 (or by the arithmetic decoding core 270 thereof), this information may naturally also bn useful for the arithmetic encoding core 370, because the arithmetic encoding core 370 should provide an encoded representation 31 1 of the binary secjuence which is decodab!e by a corresponding video decoder.
• the arithmetic encoder 320 may be configured to derive the range value 362 for the interval subdivision on the basis of the binary sequence 322 representing the video content, and also on the basis of some parameters, like the window size information n a, n b , or the bit size information BITS a , BITS b .
• the derivation of the range values 362 is substantially identical to the derivation of the range values 162, 262, except for the fact that the binary sequence 322 is evaluated instead of the decoded binary sequence 222.
• previously encoded binary values are used instead of previously decoded binary values when deriving the range values 362 on the basis of the binary sequence 322.
• the arithmetic encoder 320 of the video encoder comprises source statistic value determination blocks 330, 340, which are substantially identical to the source statistic value determination blocks 130, 140, 230, 240, except for the fact that they provide the first source statistic value 332 and the second source statistic value 342 on the basis of previously encoded binary values (rather than on the basis of previously decoded binary value).
• the source statistic value determinators 330, 340 may, for example, use the parameters n a, n b , BITS a and BITS b .
• the functionality of the source statistic determinators 330, 340 may, for example, be defined by the equations (3) and (4) as described below, or by equation (5) as described below.
• the arithmetic encoder 320 also comprises a source statistic value combination block 350, which may correspond to the source statistic value combination block 150 or the source statistic value combination block 250. Accordingly, a combined source statistic value 352 is provided, wherein the functionality of the source statistic value combination block 350 may, for example, follow equation (6) or equations (10) and (1 1 ) or equation (16).
• the arithmetic encoder 320 also comprises an optional value processing 366, which may correspond to the value processing 266 shown in Fig. 2.
• the value processing 366 may receive the combined source statistic value 352 and may, for example, provide an information 367 about a most probable binary value and/or a probability information or state index information 368.
• the arithmetic encoder 320 also comprises a mapping table 369, which may correspond to the mapping table 269 described with reference to Fig. 2. Accordingly, the one. or more range values 362 for the interval subdivision may be provided by the evaluation of the mapping table 369 using a probability value or state index 368.
• the arithmetic encoder 320 may select the range values 362 on the basis of the binary sequence 322 in the same manner as the arithmetic decoder 220 selects the range values 262 on the basis of the decoded binary sequence 222. Accordingly, synchronism between the arithmetic encoder 320 of the video encoder 300 and the arithmetic decoder 220 of the video decoder 200 can be achieved, provided that there are no decoding errors and provided that the same parameters are used at the side of the video encoder and at the side of the video decoder.
• the arithmetic encoder 320 or, generally speaking, the video encoder 300, comprises a control block 390, which may determine the parameters n a , n b , BITS a , BITS b , and any other parameters which may be necessary.
• the control blocks 390 may use some empiric mechanisms to decide about the parameters.
• the control block 390 may also analyze, which combination of parameters results in a lowest possible bit rate (or fulfils any other optimization criterion).
• the video encoder 300 may, for example, provide, within the encoded video information 312, some control information which controls the operation of the video decoder 200.
• this control information (or configuration information) may comprise one or more of the following: a slice type information, a cabac init flag, a ws_flag and a ctu_ws_flag.
• the functionality of these configuration information items, and the possible encoding and decoding thereof, will be described below.
• the slice type information may be provided like in the video encoders according to standards H.264 or FI.265 (for example, by the video binary sequence provision 380) and may be used by the video reconstruction 180,280 like in video decoders according to standards H.264 or H.265.
• the slice type information may be used in the decision about the window size(s), as outlined herein.
• distribution of the functional blocks can be modified In the embodiment of Fig. 3, the arithmetic encoding core 370, the source statistic value determinator blocks 330, 340, the source statistic value combination block 350, the value processing block 366 and the mapping table 369 and the control 390 have been described as being part of the arithmetic encoder. However, said blocks could also generally be blocks (or functionalities) of the video encoder.
• an aspect or embodiment of the invention creates a probability estimation method (or probability estimation concept) for binary arithmetic coding
• Context model update is a key feature of efficient binary arithmetic entropy coders by offering the possibility to adapt the internal state of the coder to the underlying source statistics.
• each context model is equipped with an independent probability estimation stage which provides a probability for decoding or encoding the next binary symbol (bin) that is assigned to this context model.
• the estimators described in this document are based on two counter variables that keep track of the source statistics of the assigned sequence of bins.
• Each counter (which may, for example, implement the functionality of the source statistic value determinators 130, 140, 230, 240, 330, 340) is equipped with a window size w that determines to which extent the past number of binary symbols (for example, the previously encoded or previously decoded binary values, encoded or decoded using a context model for which the source statistic values are determined) influence the count (or, generally speaking, the value of the source statistic value).
• the update is derived, for example, by the so-called virtual sliding window approach. After a bin (or binary value) with the value x t is decoded or encoded the update of the counts a t and b t (which may be considered as source statistic values) can be implemented as described in the following.
• Q A, B:
• the output of the probability estimator may be used directly (for example, to determine range values for the arithmetic encoding/decoding core) or may require further processing like mapping the value to the closest probability of a set of quantized probability values (or to one or more range values for an arithmetic encoding/decoding core).
• This update rule requires a clipping operation to be performed which guarantees that the counts remain in the valid range (0, 2 BiTSq ) exclusively :
• n q may be chosen
• x t may represent a previously encoded binary value (having, for example, values 0 or 1 ) or a previously decoded binary value.
• equations (3) and (4) may be substituted by an update calculation that distinguishes between the value of the binary symbol:
• equations (3) and (4) or equation (5) do not deliver identical bit streams due to the right-shift operation
• the operations according to equations (3) and (4) or according to equation (5) may be performed by the source statistic value determinators described herein.
• the source statistic value determinators may also perforin a different functionality (e.g. according to equation (1 )) in some embodiments.
• the weighted result c t from equation (6) (which may be considered as a combined source statistic value) can (optionally) be used to derive a probability p t (for example, by the value processing):
• an address index for example pStateldx which is dependent on the design of the arithmetic coding engine.
• the probability estimation of the finite state machine of H.265/HEVC is performed using a combination of two variables.
• the first variable pStateldx stores a state index which represents the probability for the least probable symbol (LPS) in the range (0, 0.5]
• the second variable valMPS contains the value of the most probable symbol(MPS).
• the table with the pre-ca!culated range values for the interval subdivision of the binary arithmetic coder only covers the range (0, 0.5]
• a conventional mapping table or a mapping table having a conventional structure like in a H.264 or H.265 video encoder or decoder
• mapping of the result c t from equation (6) (e.g. of the combined source statistic value) to the state index pStateldx value may, for example, be calculated by (which may, for example, include a quantization down to a resolution of a mapping table):
• the arithmetic coding engine can be modified by two options alternatively.
• Either the table containing the pre-calculated range values for the interval subdivision can be extended to cover the whole probability interval (0, 1 ) (such that, for example a table entry of the mapping table can directly be selected using the combined source statistic value or using the probability value p t (or p(0) or p( 1 ))) or the table is omitted and the range values are calculated on the fly using arithmetic (for example, on the basis of the combined source statistic vaue or on the basis of the probability value pt or p(0) or p(1 )).
• the BITS q variables can be dimensioned depending on the window size (or can be dynamically adapted according to the currently used window size).
• n a 4 and n b — 8, independent of any other parameters of the video codec.
• the window size pair for each context model is set to custom values for n (for example, for n a and/or n b ) in the range [1 ,1 1 ], inclusively, dependent on the context model, the cabac init flag and the slice type.
• the cabac init flag is part of the H.265/HEVC and is referred to as“cabacjnitjlag" [2]
• the window size values (for example, n a and n b ) for one context model can, for example, be stored in a read only memory as an 8-bit variable, with m a stored in the most significant nibble and n b stored at the least significant nibble: S pai r (m a «t 4) + 3 ⁇ 4.
• the structure of the ws pair variable as depicted in Table 1 is an example of a generous layout for software implementations, where memory requirements are not critical.
• the number of bits used to store n a and n b is capable of storing more combinations than needed.
• ws pair o,wS pair1 and ws pair2 there are three pre-defined values of ws pair , which are designated by ws pair o,wS pair1 and ws pair2 .
• One of these pre-defined values of wspair is selected (for example by the control block) on the basis of the“Slice Type” information and on the basis of a so-called“cabac init flag” (which may optionally also determine initialization values for the source statistic values and/or for other variables of the vide encoder or video decoder). For example, if the slice type is bi-predictive (“B"), the“cabac init flag" may decide which of the two pre-defined values S pair o and ws pair1 is actually used to determine n a and n b (or w a and w b ).
• the“cabac init flag” may decide which of the two pre-defined values ws pai r0 and ws pa jr 1 is actually used to determine n a and n b (or w a and w b ) (wherein an opposite allocation between the value of the cabac init flag and the pre-defined values ws pajr o and ws pa i r1 may be used when compared to the case of the bi-predictive slice type. If the slice is encoded using an independent coding mode (“Intra”), the cabac init flag may be inactive, and the pre-defined value ws pa i r2 may always be used. It should also be noted that the table 2 represents only one context model, and that different tables may be available for different context models (wherein, for example, the whole process odf range value determination may work independently for different context models).
• a context model should be initialized with the intended window sizes before decoding or encoding the first bin.
• the initialization process shall be performed at the same time when the initial probability state/count of each context model is set.
• a bin for example, a binary value
• the count is weighted, for example, as in equation (1 1 ) and used to derive a state index, for example, as defined by equations (12) and ( 13).
• a bin for example a binary value
• the counter variables or source statistic values
• the specified custom window size for example, as defined by n a and n b
• the probability estimation method (or concept) is configured identical to embodiment (a) but equation (14) is used to derive a state index.
• the state index addresses a table (for example, a mapping table for providing range values) with pre-calculated range values that cover the whole probability interval (0,1 ).
• the source statistic value determination blocks) is configured as specified in embodiment (a) but the individual counters of a context model (or the two source statistic values of the respective context model) are not equally weighted as in equation (1 1 ). Instead the weighting operation includes the weights of the individual counters
• the probability estimation method (for example, the source statistic value determination blocks) is configured as in embodiment (d) but the weights of the individual counters (for example, of the two source statistic values) are altered during the encoding or decoding process.
• the probability estimator (or the source statistic value determination blocks) is configured identical to embodiment (a), but the set of available window sizes is limited to a smaller subset to reduce memory requirements used to store the ws nir variables (for example, by adapting the control block).
• ?p,must not be smaller than n a , which for the allowed range [1 , 1 1] results in 66 valid combinations.
• the number of bits used by the ws pair variable can be reduced to a more compact layout as outlined in Table 1 - Bitmask of 8-bit variable (g)
• the probability estimator or the source statistic value determination blocks
• the number of valid combinations is reduced by omitting hand-selected ws pair variables (for example, by adapting the control block).
• the probability estimator (or the source statistic value determination blocks) is configured as specified by embodiment (f), but the number of valid combinations is reduced to ws pair variables, where the difference between n b and n a is at least 3 (n b - n a ⁇ 3) (for example, by adapting the control block).
• the probability estimation method or concept is configured as in embodiment (a) but the counters’ resolutions (or the resolution of the source statistic values) are not equal: e.g. BITS a 1 BITS (for example, by adapting the control block, or by adapting the source statistic value determination blocks).
• the counters or for source statistic values that use smaller window sizes, e.g.
• n log 2 w ⁇ 5
• the probability estimation method or concept is configured as in embodiment (j), but the sum of the BITS q variables of one context model is equal for all context models (which may, for example, be achieved by an adaptation of the control block)
• each context model requires the same number of bits and addressing specific context models can be performed by a simple offset calculation (for example, if the source statistic values associated with different context models are stored in an array or in a contiguous memory region).
• the probability estimator (for example, the source statistic value determination blocks) is configured as described in embodiment (I), but the number of valid window size combinations is reduced (for example, by an adaptation of the control block), because of the constraint that the sum of the BlTS q variables is equal for all context models.
• the probability estimator for example, the source statistic value determination blocks
• the window size of one counter is set to infinite.
• the count is never updated and the complexity can be reduced in terms of arithmetic operations and memory consumption.
• one of the source statistic values can be replaced by a fixed value. Accordingly, the combined source statistic value can be obtained by combination of one of the source statistic values with said fixed value.
• the probability estimator (for example, the source statistic value determination blocks) is configured as described in embodiment (a) but the context models are initialized dependent on a separate flag that is transmitted for each slice (for example, provided by a video encoder and evaluated by a video decoder). Instead of initializing the window sizes dependent on the slice type and the cabac init flag, a separate flag“ws_flag” is introduced which specifies the set used (for example, the values of n a and n b used). As a result, the assignment of Table 2 is extended as shown in Table 3.
• control block may, for example, evaluate the ws flag and adjust the window sizes in dependence on the ws flag (and typically also in dependence on the slice type parameter, which is typically also included in the video bitstream as a control information or configuration information, for example, for each slice or for selected slices)
• the probability estimator for example, the source statistic value determination blocks
• the initialization of window sizes is additionally dependent on the temporal level of the current slice.
• the control block or the video encoder or decoder
• the probability estimator for example, the source statistic value determination blocks
• the initialization of window sizes is additionally dependent on the quantization parameter of the current slice.
• the control block or the video encoder or decoder
• the control block may, for example, be configured to consider the quantization parameter of the current slice when determining the window sizes.
• the probability estimator (for example, the source statistic value determination blocks) is configured as described in embodiment (p) but the initialization is additionally dependent on the quantization parameter.
• a ws pair variable is specified for each item of the product space of temporal level, quantization parameter, slice type and cabac init flag.
• the control block (or the video encoder or decoder) may, for example, be configured to consider the temporal level and the quantization parameter of the current slice when determining the window sizes.
• the probability estimation method or concept (for example, the source statistic value determination blocks) is configured identical to embodiment (a) but the window size variables are altered during the decoding or encoding process (for example, within a slice and/or between different slices and/or between different frames) , dependent on the position.
• the window sisos (for e v arnplo, ro and n. , of bold counters (or source statistic value cloterrninafors i is set to:
• this may mean a selective reduction of the window size if the current bin originates from a CTUI that is part of the first row or of the first column of a slice.
• the window sizes are set to their default values (for example, as defined on the basis of the slice type and any other parameters or flags which may be considered, for example, according to one of tables 2 and 3).
• the probability estimator may adapt faster or slower to the underlying statistics of the source bin sequence in the beginning of a slice or a row of CTUs.
• the probability estimation method or concept (for example, the source statistic value determination blocks) is configured identical to embodiment (a) but an additional flag is used (for example, provided by a video encoder and evaluated by a video decoder) to alter the window size during decoding or encoding of a slice.
• an additional flag is used (for example, provided by a video encoder and evaluated by a video decoder) to alter the window size during decoding or encoding of a slice.
• there is a flag“ctu_ws__flag” for example, for every CTU which is used to increase or decrease the n a and n b variables for all context models or a subset (for example, of the context models).
• the“ctu_ws_flag” is signalled, for example, before every CTU and during decoding the“ctu_ws_flag” is read, for example, before every CTU.
• n a and n b variables of all context models are increased or decreased by one if the result is still in the valid range [1 ,1 1].
• the probability estimation method or concept (for example, the source statistic value determination blocks) is configured identical to embodiment (t) but only the window sizes of a subset of all context models are altered, dependent on the ctu_ws_flag.
• FIG. 4 shows a flowchart of a m H;I 400, according to on rnboclirneni oi ihe invention
• Fig 5 shows a flowchart of a method 500, according to an embodiment of the invention.
• Fig. 6 shows a schematic (simplified) representation of a video stream (bitstream), according to an embodiment of the present invention.
• video stream 600 can optionally be supplemented by any of the features and details described herein, both individually and taken in combination. 8).
• document in [4] describes custom window sizes defined for each context model.
• the window sizes are not derived dependent on the slice type or the cabac init flag.
• Two pairs of window sizes are defined for each context model and the quantization parameter is used to select one pair.
• a different range of window sizes is provided.
• the document in [5] describes a method which can be compared as an alternative to the fast attack mode described in claim (s).
• the document describes that an additional pair of window sizes is defined for the first 64 bins assigned the context model. After the first 64 bins are decoded or encoded the window size are set to their default value. This approach requires a separate counter variable which tracks the number of bins coded for each context model.
• aspects have been described in the context of an apparatus, it is clear that these aspects also represent a description of the corresponding method, where a block or device corresponds to a method step or a feature of a method step. Analogously, aspects described in the context of a method step also represent a description of a corresponding block or item or feature of a corresponding apparatus.
• Some or all of the method steps may be executed by (or using) a hardware apparatus, like for example, a microprocessor, a programmable computer or an electronic circuit.
• iho moot important method Stops may bo executed by such an apparatus
• the inventive encoded video signal (or video stream or video bitstream) can be stored on a digital storage medium or can be transmitted on a transmission medium such as a wireless transmission medium or a wired transmission medium such as the Internet.
• embodiments of the invention can be implemented in hardware or in software.
• the implementation can be performed using a digital storage medium, for example a floppy disk, a DVD, a Blu-Ray, a CD, a ROM, a PROM, an EPROM, an EEPROM or a FLASH memory, having electronically readable control signals stored thereon, which cooperate (or are capable of cooperating) with a programmable computer system such that the respective method is performed. Therefore, the digital storage medium may be computer readable.
• Some embodiments according to the invention comprise a data carrier having electronically readable control signals, which are capable of cooperating with a programmable computer system, such that one of the methods described herein is performed.
• embodiments of the present invention can be implemented as a computer program product with a program code, the program code being operative for performing one of the methods when the computer program product runs on a computer.
• the program code may for example be stored on a machine readable carrier.
• inventions comprise the computer program for performing one of the methods described herein, stored on a machine readable carrier.
• an embodiment of the inventive method is, therefore, a computer program having a program code for performing one of the methods described herein, when the computer program runs on a computer.
• a further embodiment of the inventive methods is. therefore, a data carrier (or a digital storage medium, or a computer-readable medium) comprising, recorded thereon, the computer program for performing one of the methods described herein.
• the data earner, the digital storage medium or the recorded medium are typically tangible and/or non
• a further embodiment of the inventive method is, therefore, a data stream or a sequence of signals representing the computer program for performing one of the methods described herein.
• the data stream or the sequence of signals may for example be configured to be transferred via a data communication connection, for example via the Internet.
• a further embodiment comprises a processing means, for example a computer, or a programmable logic device, configured to or adapted to perform one of the methods described herein.
• a processing means for example a computer, or a programmable logic device, configured to or adapted to perform one of the methods described herein.
• a further embodiment comprises a computer having installed thereon the computer program for performing one of the methods described herein.
• a further embodiment according to the invention comprises an apparatus or a system configured to transfer (for example, electronically or optically) a computer program for performing one of the methods described herein to a receiver.
• the receiver may, for example, be a computer, a mobile device, a memory device or the like.
• the apparatus or system may, for example, comprise a file server for transferring the computer program to the receiver.
• a programmable logic device for example a field programmable gate array
• a field programmable gate array may cooperate with a microprocessor in order to perform one of the methods described herein.
• the methods are preferably performed by any hardware apparatus.
• the apparatus described herein may be implemented using a hardware apparatus, or using a computer, or using a combination of a hardware apparatus and a computer.
• the apparatus described herein, or any components of the apparatus described herein may be implemented at least partially in hardware and/or in software.
• the methods described herein may be performed using a hardware apparatus, or using a computer, or using a combination of a hardware apparatus and a computer.
• Arithmetic coding with progressive context-dependent double-window adaptation response http://phenix.it- sudparis .eu/jvet/doc en d use r/documents/8 M a cau/wq11/JVET - H0067-v2.zip

Applications Claiming Priority (2)

Publications (1)

Family

ID=61868360

Family Applications (1)

Country Status (7)

Families Citing this family (6)

Family Cites Families (3)

• 2019
  • 2019-03-29 TW TW108111309A patent/TWI721396B/zh active
  • 2019-03-29 KR KR1020237043745A patent/KR20240005114A/ko active Application Filing
  • 2019-03-29 WO PCT/EP2019/058057 patent/WO2019185893A1/en active Application Filing
  • 2019-03-29 KR KR1020237012824A patent/KR102617187B1/ko active IP Right Grant
  • 2019-03-29 KR KR1020207031124A patent/KR102523750B1/ko active IP Right Grant
  • 2019-03-29 JP JP2020552198A patent/JP7181941B2/ja active Active
  • 2019-03-29 EP EP19713070.1A patent/EP3777197A1/en active Pending
  • 2019-03-29 TW TW110103575A patent/TWI799785B/zh active
  • 2019-03-29 CN CN201980036029.6A patent/CN112313960B/zh active Active
• 2020
  • 2020-09-29 US US17/036,777 patent/US11457230B2/en active Active
• 2022
  • 2022-08-31 US US17/900,402 patent/US20220417538A1/en active Pending
  • 2022-11-18 JP JP2022184990A patent/JP7451655B2/ja active Active

Also Published As

Similar Documents

Legal Events